A Novel Template Matching Method Incorporating a Multi‐Candidate Region Optimization Strategy for the Initial Localization of Mars Helicopter

ABSTRACT The application of Unmanned Aerial Vehicles (UAVs) on Mars marks a significant advancement in planetary exploration missions. In the Mars 2020 mission, NASA successfully operated a helicopter on the Martian surface and performed multiple flight experiments. A critical challenge in maximizing the scientific yield of the Mars Helicopter is achieving high‐precision localization. To address this localization issue in weakly textured Martian environments, this paper proposes a novel template matching method that incorporates a multi‐candidate region optimization strategy to improve the accuracy of the initial localization of the Mars Helicopter. Given the UAV images and the reference orbital DOM images, a multi‐candidate region extraction method is first designed to identify the potential candidate regions in the DOM corresponding to the UAV image. Subsequently, a candidate region ranking operation is performed to rearrange the generated multi‐candidate regions to determine the optimal match. We validated our method with real flight data of the Ingenuity helicopter from the NASA Mars 2020 project and conducted a comparative analysis against the state‐of‐the‐art techniques. Experimental results show that our method achieves better results compared with existing methods, which demonstrates the potential for application in engineering tasks.